1. 43
  1. 22

    AMP is one of those things that really shocked me when I found out that it wasn’t satire.

    1. 14

      “At a scale of billions of users, this has the effect of further reinforcing Google’s dominance of the Web.”

      You say this like it was an accident.

      1. 3

        Good start… now what about templating alternatives like Telegram’s Instant View? Millions of links being sent over IM every day are being rewritten into new templates created by a third party. Sure, it addresses the speed / mobile accessibility concerns, but it’s also very heavy-handed backend processing that’s a black box to users.

        1. 3

          We are currently running a $315,000 competition to create Instant View templates for news websites and blogs. Everyone is welcome to participate.

          That’s a lot of $s for something which is run by… who exactly again?

          1. 1

            Google AMP doesn’t exist to solve backend issues, though. That’s a separate concern, I’d say.

            1. 1

              Before clicking your link, I was unfamiliar with Telegram and Instant View, so take this with a grain of salt…

              Isn’t this just another version of the same thing? Doesn’t an Instant View prevent clicks to the host domain?

              And, conversely, isn’t this all a consequence of business models riddled with terrible ad networks? Dropping AMP or Instant View or any other scraper/viewer doesn’t fix that weakness.

              1. 2

                More like mix of Facebook’s Instant Articles and Reader Mode from Safari/Firebox (also Pocket). It doesn’t require special markup to put on webpage but it has crowdsourced rewrite rules that remove cruft from webpages. It loads processed webpages from their server though, unlike Reader modes.

                At least Telegram leaves links posted to chat as is, looking like links, with underline, leading to original URL, and adds “Instant view” button alongside, which looks like button and opens instant article popup.

                BTW, their rule language is crappy.

                1. 1

                  Telegram is/was important in the recent Iran riots.

                  For example: https://www.politico.com/magazine/story/2018/01/01/irans-telegram-revolution-216206

                  Unlike Twitter, millions of Iranians use Telegram in their everyday lives—around 40 million monthly users in a country of 45 million overall online users, according to the latest ITU statistics.

                  1. 1

                    To my understanding it is - just a different content/service provider that wants to create a platform to give the same experience to its users. There are some differences in the implementation by telegram and the effort that needs to be put by the sites developers, but it is still served by cache that is kept on their servers. In this point you’re getting a good experience loading, say, Medium articles in Telegram - but you don’t get the same experience outside of it.

                2. 2

                  Having looked over AMP and seeing all the different points it does not seem as bad as people make it out to be, at least from a technical point of view.

                  These are the two main points of contention.

                  1. Content that “opts in” to AMP and the associated hosting within Google’s domain is granted preferential search promotion, including (for news articles) a position above all other results.
                  2. When a user navigates from Google to a piece of content Google has recommended, they are, unwittingly, remaining within Google’s ecosystem.

                  The first does not bother me so much. It is a standardized way of limiting page size growth, and enforcing best practices for speed. It is a formal way to enforce getting 90 on a webpage speed test. Google has used speed as a ranking factor for awhile now.

                  The second point does have some issue. Originally the address of the page was obscured when it was served from Google’s servers, and that, in my opinion, is not great. This has now been, somewhat, resolved. While it still connects the user and the publisher, Google is overstepping it’s middleman role.

                  I don’t like the idea of Google hosting cached versions of my content, but I can live with it.

                  Will someone expand on the AMP cache, and explain why it is a more horrible thing than I currently understand?

                  1. 1

                    Re: the AMP cache, there might be other reasons, but I think it creates an unfair playing field. It optimizes for people who have teams to rewrite/add to their templating code to fit AMP when there very well could be better more relevant content that is just managed by 1 person or non-tech people who would need to outsource.

                    1. 1

                      I see where that could be the case in some situations, but let me play devil’s advocate for a moment and say; AMP makes it easier for a single person to optimize a website. As a specific set of rules, it is easier to create layouts, and the validation tools help guide one to a speedy websites. Using the specific set of rules is more straightforward than having to know a ton of micro-optimizations, and how they all need to fit together.

                      I just added AMP to my website. I based the AMP pages on my existing layout, and only had to change a few things. All together it took me about three days.

                      But I came here to share this:

                      TL;DR: We are making changes to how AMP works in platforms such as Google Search that will enable linked pages to appear under publishers’ URLs instead of the google.com/amp URL space while maintaining the performance and privacy benefits of AMP Cache serving.


                  2. 2

                    Completely agree that AMP is a Real Bad Thing, but letters of complaint like this seem completely toothless. Kind of sad. Especially in context of the Title II / net neutrality hoopla in the US recently, which got much more PR than I think we can ever expect for this issue.

                    Anyone care to speculate on the prospects of an antitrust suit? Or any other form of effective regulatory action? Or technical solutions?